Composition and Decomposition in Artistic Portraits, Scientific Photography, and Deepfake Videos

This paper aims to describe the specificity of the representational devices which, over the centuries, have given prominence to the face. I will study three representational devices: the classical painted portrait, the composite photography pioneered by Francis Galton at the end of the 19th century and the representation of the face in contemporary deepfake videos using computational methods on big data. I will take account of the different framing strategies and uses of the face employed by artistic painting, scientific photography, and deepfake videos in order to identify the representational rules as well as the models of individuality/generality that have succeeded one another. The main questions addressed will be: How to represent a specific identity? How, on the other hand, to represent an average or typical identity or behavior of a specific people? And finally: How to construct, through multiple images, a typical identity of oneself, in order to possibly make it substitutable for someone else’s identity?